release 2016.08.22

Helper functions should have consistent default values

.github/ISSUE_TEMPLATE.md

@@ -6,8 +6,8 @@
 
 ---
 
-### Make sure you are using the *latest* version: run `youtube-dl --version` and ensure your version is *2016.08.19*. If it's not read [this FAQ entry](https://github.com/rg3/youtube-dl/blob/master/README.md#how-do-i-update-youtube-dl) and update. Issues with outdated version will be rejected.
+### Make sure you are using the *latest* version: run `youtube-dl --version` and ensure your version is *2016.08.22*. If it's not read [this FAQ entry](https://github.com/rg3/youtube-dl/blob/master/README.md#how-do-i-update-youtube-dl) and update. Issues with outdated version will be rejected.
-- [ ] I've **verified** and **I assure** that I'm running youtube-dl **2016.08.19**
+- [ ] I've **verified** and **I assure** that I'm running youtube-dl **2016.08.22**
 
 ### Before submitting an *issue* make sure you have:
 - [ ] At least skimmed through [README](https://github.com/rg3/youtube-dl/blob/master/README.md) and **most notably** [FAQ](https://github.com/rg3/youtube-dl#faq) and [BUGS](https://github.com/rg3/youtube-dl#bugs) sections
@@ -35,7 +35,7 @@ $ youtube-dl -v <your command line>
 [debug] User config: []
 [debug] Command-line args: [u'-v', u'http://www.youtube.com/watch?v=BaW_jenozKcj']
 [debug] Encodings: locale cp1251, fs mbcs, out cp866, pref cp1251
-[debug] youtube-dl version 2016.08.19
+[debug] youtube-dl version 2016.08.22
 [debug] Python version 2.7.11 - Windows-2003Server-5.2.3790-SP2
 [debug] exe versions: ffmpeg N-75573-g1d0487f, ffprobe N-75573-g1d0487f, rtmpdump 2.4
 [debug] Proxy map: {}

ChangeLog

@@ -1,3 +1,30 @@
+version 2016.08.22
+
+Core
+* Improve formats and subtitles extension auto calculation
++ Recognize full unit names in parse_filesize
++ Add support for m3u8 manifests in HTML5 multimedia tags
+* Fix octal/hexadecimal number detection in js_to_json
+
+Extractors
++ [ivi] Add support for 720p and 1080p
++ [charlierose] Add new extractor (#10382)
+* [1tv] Fix extraction (#9249)
+* [twitch] Renew authentication
+* [kaltura] Improve subtitles extension calculation
++ [zingmp3] Add support for video clips
+* [zingmp3] Fix extraction (#10041)
+* [kaltura] Improve subtitles extraction (#10279)
+* [cultureunplugged] Fix extraction (#10330)
++ [cnn] Add support for money.cnn.com (#2797)
+* [cbsnews] Fix extraction (#10362)
+* [cbs] Fix extraction (#10393)
++ [litv] Support 'promo' URLs (#10385)
+* [snotr] Fix extraction (#10338)
+* [n-tv.de] Fix extraction (#10331)
+* [globo:article] Relax URL and video id regular expressions (#10379)
+
+
 version 2016.08.19
 
 Core

README.md

@@ -645,7 +645,11 @@ $ youtube-dl -f 'best[filesize<50M]'
 
 # Download best format available via direct link over HTTP/HTTPS protocol
 $ youtube-dl -f '(bestvideo+bestaudio/best)[protocol^=http]'
+
+# Download the best video format and the best audio format without merging them
+$ youtube-dl -f 'bestvideo,bestaudio' -o '%(title)s.f%(format_id)s.%(ext)s'
 ```
+Note that in the last example, an output template is recommended as bestvideo and bestaudio may have the same file name.
 
 
 # VIDEO SELECTION

docs/supportedsites.md

@@ -121,6 +121,7 @@
  - **CDA**
  - **CeskaTelevize**
  - **channel9**: Channel 9
+ - **CharlieRose**
  - **Chaturbate**
  - **Chilloutzone**
  - **chirbit**
@@ -893,5 +894,4 @@
  - **Zapiks**
  - **ZDF**
  - **ZDFChannel**
- - **zingmp3:album**: mp3.zing.vn albums
+ - **zingmp3**: mp3.zing.vn
- - **zingmp3:song**: mp3.zing.vn songs

test/test_utils.py

@@ -712,6 +712,9 @@ class TestUtil(unittest.TestCase):
         inp = '''{"foo":101}'''
         self.assertEqual(js_to_json(inp), '''{"foo":101}''')
 
+        inp = '''{"duration": "00:01:07"}'''
+        self.assertEqual(js_to_json(inp), '''{"duration": "00:01:07"}''')
+
     def test_js_to_json_edgecases(self):
         on = js_to_json("{abc_def:'1\\'\\\\2\\\\\\'3\"4'}")
         self.assertEqual(json.loads(on), {"abc_def": "1'\\2\\'3\"4"})
@@ -820,6 +823,7 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(parse_filesize('1.2tb'), 1200000000000)
         self.assertEqual(parse_filesize('1,24 KB'), 1240)
         self.assertEqual(parse_filesize('1,24 kb'), 1240)
+        self.assertEqual(parse_filesize('8.5 megabytes'), 8500000)
 
     def test_parse_count(self):
         self.assertEqual(parse_count(None), None)

youtube_dl/YoutubeDL.py

@@ -1299,7 +1299,7 @@ class YoutubeDL(object):
                 for subtitle_format in subtitle:
                     if subtitle_format.get('url'):
                         subtitle_format['url'] = sanitize_url(subtitle_format['url'])
-                    if 'ext' not in subtitle_format:
+                    if subtitle_format.get('ext') is None:
                         subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
 
         if self.params.get('listsubtitles', False):
@@ -1354,7 +1354,7 @@ class YoutubeDL(object):
                     note=' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '',
                 )
             # Automatically determine file extension if missing
-            if 'ext' not in format:
+            if format.get('ext') is None:
                 format['ext'] = determine_ext(format['url']).lower()
             # Automatically determine protocol if missing (useful for format
             # selection purposes)

youtube_dl/extractor/cbs.py

@@ -4,6 +4,7 @@ from .theplatform import ThePlatformFeedIE
 from ..utils import (
     int_or_none,
     find_xpath_attr,
+    ExtractorError,
 )
 
 
@@ -17,19 +18,6 @@ class CBSBaseIE(ThePlatformFeedIE):
             }]
         } if closed_caption_e is not None and closed_caption_e.attrib.get('value') else []
 
-    def _extract_video_info(self, filter_query, video_id):
-        return self._extract_feed_info(
-            'dJ5BDC', 'VxxJg8Ymh8sE', filter_query, video_id, lambda entry: {
-                'series': entry.get('cbs$SeriesTitle'),
-                'season_number': int_or_none(entry.get('cbs$SeasonNumber')),
-                'episode': entry.get('cbs$EpisodeTitle'),
-                'episode_number': int_or_none(entry.get('cbs$EpisodeNumber')),
-            }, {
-                'StreamPack': {
-                    'manifest': 'm3u',
-                }
-            })
-
 
 class CBSIE(CBSBaseIE):
     _VALID_URL = r'(?:cbs:|https?://(?:www\.)?(?:cbs\.com/shows/[^/]+/video|colbertlateshow\.com/(?:video|podcasts))/)(?P<id>[\w-]+)'
@@ -38,7 +26,6 @@ class CBSIE(CBSBaseIE):
         'url': 'http://www.cbs.com/shows/garth-brooks/video/_u7W953k6la293J7EPTd9oHkSPs6Xn6_/connect-chat-feat-garth-brooks/',
         'info_dict': {
             'id': '_u7W953k6la293J7EPTd9oHkSPs6Xn6_',
-            'display_id': 'connect-chat-feat-garth-brooks',
             'ext': 'mp4',
             'title': 'Connect Chat feat. Garth Brooks',
             'description': 'Connect with country music singer Garth Brooks, as he chats with fans on Wednesday November 27, 2013. Be sure to tune in to Garth Brooks: Live from Las Vegas, Friday November 29, at 9/8c on CBS!',
@@ -47,7 +34,10 @@ class CBSIE(CBSBaseIE):
             'upload_date': '20131127',
             'uploader': 'CBSI-NEW',
         },
-        'expected_warnings': ['Failed to download m3u8 information'],
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
         '_skip': 'Blocked outside the US',
     }, {
         'url': 'http://colbertlateshow.com/video/8GmB0oY0McANFvp2aEffk9jZZZ2YyXxy/the-colbeard/',
@@ -56,8 +46,31 @@ class CBSIE(CBSBaseIE):
         'url': 'http://www.colbertlateshow.com/podcasts/dYSwjqPs_X1tvbV_P2FcPWRa_qT6akTC/in-the-bad-room-with-stephen/',
         'only_matching': True,
     }]
-    TP_RELEASE_URL_TEMPLATE = 'http://link.theplatform.com/s/dJ5BDC/%s?mbr=true'
+
+    def _extract_video_info(self, guid):
+        path = 'dJ5BDC/media/guid/2198311517/' + guid
+        smil_url = 'http://link.theplatform.com/s/%s?mbr=true' % path
+        formats, subtitles = self._extract_theplatform_smil(smil_url + '&manifest=m3u', guid)
+        for r in ('HLS&formats=M3U', 'RTMP', 'WIFI', '3G'):
+            try:
+                tp_formats, _ = self._extract_theplatform_smil(smil_url + '&assetTypes=' + r, guid, 'Downloading %s SMIL data' % r.split('&')[0])
+                formats.extend(tp_formats)
+            except ExtractorError:
+                continue
+        self._sort_formats(formats)
+        metadata = self._download_theplatform_metadata(path, guid)
+        info = self._parse_theplatform_metadata(metadata)
+        info.update({
+            'id': guid,
+            'formats': formats,
+            'subtitles': subtitles,
+            'series': metadata.get('cbs$SeriesTitle'),
+            'season_number': int_or_none(metadata.get('cbs$SeasonNumber')),
+            'episode': metadata.get('cbs$EpisodeTitle'),
+            'episode_number': int_or_none(metadata.get('cbs$EpisodeNumber')),
+        })
+        return info
 
     def _real_extract(self, url):
         content_id = self._match_id(url)
-        return self._extract_video_info('byGuid=%s' % content_id, content_id)
+        return self._extract_video_info(content_id)

youtube_dl/extractor/cbsnews.py

@@ -2,13 +2,13 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from .cbs import CBSBaseIE
+from .cbs import CBSIE
 from ..utils import (
     parse_duration,
 )
 
 
-class CBSNewsIE(CBSBaseIE):
+class CBSNewsIE(CBSIE):
     IE_DESC = 'CBS News'
     _VALID_URL = r'https?://(?:www\.)?cbsnews\.com/(?:news|videos)/(?P<id>[\da-z_-]+)'
 
@@ -35,7 +35,8 @@ class CBSNewsIE(CBSBaseIE):
                 'ext': 'mp4',
                 'title': 'Fort Hood shooting: Army downplays mental illness as cause of attack',
                 'description': 'md5:4a6983e480542d8b333a947bfc64ddc7',
-                'upload_date': '19700101',
+                'upload_date': '20140404',
+                'timestamp': 1396650660,
                 'uploader': 'CBSI-NEW',
                 'thumbnail': 're:^https?://.*\.jpg$',
                 'duration': 205,
@@ -63,7 +64,7 @@ class CBSNewsIE(CBSBaseIE):
 
         item = video_info['item'] if 'item' in video_info else video_info
         guid = item['mpxRefId']
-        return self._extract_video_info('byGuid=%s' % guid, guid)
+        return self._extract_video_info(guid)
 
 
 class CBSNewsLiveVideoIE(InfoExtractor):

youtube_dl/extractor/cbssports.py

@@ -23,6 +23,9 @@ class CBSSportsIE(CBSBaseIE):
         }
     }]
 
+    def _extract_video_info(self, filter_query, video_id):
+        return self._extract_feed_info('dJ5BDC', 'VxxJg8Ymh8sE', filter_query, video_id)
+
     def _real_extract(self, url):
         video_id = self._match_id(url)
         return self._extract_video_info('byId=%s' % video_id, video_id)

youtube_dl/extractor/charlierose.py

@@ -0,0 +1,51 @@
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import remove_end
+
+
+class CharlieRoseIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?charlierose\.com/video(?:s|/player)/(?P<id>\d+)'
+    _TESTS = [{
+        'url': 'https://charlierose.com/videos/27996',
+        'md5': 'fda41d49e67d4ce7c2411fd2c4702e09',
+        'info_dict': {
+            'id': '27996',
+            'ext': 'mp4',
+            'title': 'Remembering Zaha Hadid',
+            'thumbnail': 're:^https?://.*\.jpg\?\d+',
+            'description': 'We revisit past conversations with Zaha Hadid, in memory of the world renowned Iraqi architect.',
+            'subtitles': {
+                'en': [{
+                    'ext': 'vtt',
+                }],
+            },
+        },
+    }, {
+        'url': 'https://charlierose.com/videos/27996',
+        'only_matching': True,
+    }]
+
+    _PLAYER_BASE = 'https://charlierose.com/video/player/%s'
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(self._PLAYER_BASE % video_id, video_id)
+
+        title = remove_end(self._og_search_title(webpage), ' - Charlie Rose')
+
+        info_dict = self._parse_html5_media_entries(
+            self._PLAYER_BASE % video_id, webpage, video_id,
+            m3u8_entry_protocol='m3u8_native')[0]
+
+        self._sort_formats(info_dict['formats'])
+        self._remove_duplicate_formats(info_dict['formats'])
+
+        info_dict.update({
+            'id': video_id,
+            'title': title,
+            'thumbnail': self._og_search_thumbnail(webpage),
+            'description': self._og_search_description(webpage),
+        })
+
+        return info_dict

youtube_dl/extractor/cnn.py

@@ -11,7 +11,7 @@ from ..utils import (
 
 
 class CNNIE(InfoExtractor):
-    _VALID_URL = r'''(?x)https?://(?:(?:edition|www)\.)?cnn\.com/video/(?:data/.+?|\?)/
+    _VALID_URL = r'''(?x)https?://(?:(?P<sub_domain>edition|www|money)\.)?cnn\.com/(?:video/(?:data/.+?|\?)/)?videos?/
         (?P<path>.+?/(?P<title>[^/]+?)(?:\.(?:[a-z\-]+)|(?=&)))'''
 
     _TESTS = [{
@@ -45,19 +45,46 @@ class CNNIE(InfoExtractor):
             'description': 'md5:e7223a503315c9f150acac52e76de086',
             'upload_date': '20141222',
         }
+    }, {
+        'url': 'http://money.cnn.com/video/news/2016/08/19/netflix-stunning-stats.cnnmoney/index.html',
+        'md5': '52a515dc1b0f001cd82e4ceda32be9d1',
+        'info_dict': {
+            'id': '/video/news/2016/08/19/netflix-stunning-stats.cnnmoney',
+            'ext': 'mp4',
+            'title': '5 stunning stats about Netflix',
+            'description': 'Did you know that Netflix has more than 80 million members? Here are five facts about the online video distributor that you probably didn\'t know.',
+            'upload_date': '20160819',
+        }
     }, {
         'url': 'http://cnn.com/video/?/video/politics/2015/03/27/pkg-arizona-senator-church-attendance-mandatory.ktvk',
         'only_matching': True,
     }, {
         'url': 'http://cnn.com/video/?/video/us/2015/04/06/dnt-baker-refuses-anti-gay-order.wkmg',
         'only_matching': True,
+    }, {
+        'url': 'http://edition.cnn.com/videos/arts/2016/04/21/olympic-games-cultural-a-z-brazil.cnn',
+        'only_matching': True,
     }]
 
+    _CONFIG = {
+        # http://edition.cnn.com/.element/apps/cvp/3.0/cfg/spider/cnn/expansion/config.xml
+        'edition': {
+            'data_src': 'http://edition.cnn.com/video/data/3.0/video/%s/index.xml',
+            'media_src': 'http://pmd.cdn.turner.com/cnn/big',
+        },
+        # http://money.cnn.com/.element/apps/cvp2/cfg/config.xml
+        'money': {
+            'data_src': 'http://money.cnn.com/video/data/4.0/video/%s.xml',
+            'media_src': 'http://ht3.cdn.turner.com/money/big',
+        },
+    }
+
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        sub_domain, path, page_title = re.match(self._VALID_URL, url).groups()
-        path = mobj.group('path')
+        if sub_domain not in ('money', 'edition'):
-        page_title = mobj.group('title')
+            sub_domain = 'edition'
-        info_url = 'http://edition.cnn.com/video/data/3.0/%s/index.xml' % path
+        config = self._CONFIG[sub_domain]
+        info_url = config['data_src'] % path
         info = self._download_xml(info_url, page_title)
 
         formats = []
@@ -66,7 +93,7 @@ class CNNIE(InfoExtractor):
             (?:_(?P<bitrate>[0-9]+)k)?
         ''')
         for f in info.findall('files/file'):
-            video_url = 'http://ht.cdn.turner.com/cnn/big%s' % (f.text.strip())
+            video_url = config['media_src'] + f.text.strip()
             fdct = {
                 'format_id': f.attrib['bitrate'],
                 'url': video_url,
@@ -146,7 +173,7 @@ class CNNBlogsIE(InfoExtractor):
 
 
 class CNNArticleIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:(?:edition|www)\.)?cnn\.com/(?!video/)'
+    _VALID_URL = r'https?://(?:(?:edition|www)\.)?cnn\.com/(?!videos?/)'
     _TEST = {
         'url': 'http://www.cnn.com/2014/12/21/politics/obama-north-koreas-hack-not-war-but-cyber-vandalism/',
         'md5': '689034c2a3d9c6dc4aa72d65a81efd01',

youtube_dl/extractor/common.py

@@ -1695,7 +1695,7 @@ class InfoExtractor(object):
                         self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
         return formats
 
-    def _parse_html5_media_entries(self, base_url, webpage):
+    def _parse_html5_media_entries(self, base_url, webpage, video_id, m3u8_id=None, m3u8_entry_protocol='m3u8'):
         def absolute_url(video_url):
             return compat_urlparse.urljoin(base_url, video_url)
 
@@ -1710,6 +1710,21 @@ class InfoExtractor(object):
                 return f
             return {}
 
+        def _media_formats(src, cur_media_type):
+            full_url = absolute_url(src)
+            if determine_ext(full_url) == 'm3u8':
+                is_plain_url = False
+                formats = self._extract_m3u8_formats(
+                    full_url, video_id, ext='mp4',
+                    entry_protocol=m3u8_entry_protocol, m3u8_id=m3u8_id)
+            else:
+                is_plain_url = True
+                formats = [{
+                    'url': full_url,
+                    'vcodec': 'none' if cur_media_type == 'audio' else None,
+                }]
+            return is_plain_url, formats
+
         entries = []
         for media_tag, media_type, media_content in re.findall(r'(?s)(<(?P<tag>video|audio)[^>]*>)(.*?)</(?P=tag)>', webpage):
             media_info = {
@@ -1719,10 +1734,8 @@ class InfoExtractor(object):
             media_attributes = extract_attributes(media_tag)
             src = media_attributes.get('src')
             if src:
-                media_info['formats'].append({
+                _, formats = _media_formats(src)
-                    'url': absolute_url(src),
+                media_info['formats'].extend(formats)
-                    'vcodec': 'none' if media_type == 'audio' else None,
-                })
             media_info['thumbnail'] = media_attributes.get('poster')
             if media_content:
                 for source_tag in re.findall(r'<source[^>]+>', media_content):
@@ -1730,12 +1743,13 @@ class InfoExtractor(object):
                     src = source_attributes.get('src')
                     if not src:
                         continue
-                    f = parse_content_type(source_attributes.get('type'))
+                    is_plain_url, formats = _media_formats(src, media_type)
-                    f.update({
+                    if is_plain_url:
-                        'url': absolute_url(src),
+                        f = parse_content_type(source_attributes.get('type'))
-                        'vcodec': 'none' if media_type == 'audio' else None,
+                        f.update(formats[0])
-                    })
+                        media_info['formats'].append(f)
-                    media_info['formats'].append(f)
+                    else:
+                        media_info['formats'].extend(formats)
                 for track_tag in re.findall(r'<track[^>]+>', media_content):
                     track_attributes = extract_attributes(track_tag)
                     kind = track_attributes.get('kind')

youtube_dl/extractor/cultureunplugged.py

@@ -1,9 +1,13 @@
 from __future__ import unicode_literals
 
 import re
+import time
 
 from .common import InfoExtractor
-from ..utils import int_or_none
+from ..utils import (
+    int_or_none,
+    HEADRequest,
+)
 
 
 class CultureUnpluggedIE(InfoExtractor):
@@ -32,6 +36,9 @@ class CultureUnpluggedIE(InfoExtractor):
         video_id = mobj.group('id')
         display_id = mobj.group('display_id') or video_id
 
+        # request setClientTimezone.php to get PHPSESSID cookie which is need to get valid json data in the next request
+        self._request_webpage(HEADRequest(
+            'http://www.cultureunplugged.com/setClientTimezone.php?timeOffset=%d' % -(time.timezone / 3600)), display_id)
         movie_data = self._download_json(
             'http://www.cultureunplugged.com/movie-data/cu-%s.json' % video_id, display_id)
 

youtube_dl/extractor/dotsub.py

@@ -10,18 +10,18 @@ from ..utils import (
 class DotsubIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?dotsub\.com/view/(?P<id>[^/]+)'
     _TEST = {
-        'url': 'http://dotsub.com/view/aed3b8b2-1889-4df5-ae63-ad85f5572f27',
+        'url': 'https://dotsub.com/view/9c63db2a-fa95-4838-8e6e-13deafe47f09',
-        'md5': '0914d4d69605090f623b7ac329fea66e',
+        'md5': '21c7ff600f545358134fea762a6d42b6',
         'info_dict': {
-            'id': 'aed3b8b2-1889-4df5-ae63-ad85f5572f27',
+            'id': '9c63db2a-fa95-4838-8e6e-13deafe47f09',
             'ext': 'flv',
-            'title': 'Pyramids of Waste (2010), AKA The Lightbulb Conspiracy - Planned obsolescence documentary',
+            'title': 'MOTIVATION - "It\'s Possible" Best Inspirational Video Ever',
-            'description': 'md5:699a0f7f50aeec6042cb3b1db2d0d074',
+            'description': 'md5:41af1e273edbbdfe4e216a78b9d34ac6',
-            'thumbnail': 're:^https?://dotsub.com/media/aed3b8b2-1889-4df5-ae63-ad85f5572f27/p',
+            'thumbnail': 're:^https?://dotsub.com/media/9c63db2a-fa95-4838-8e6e-13deafe47f09/p',
-            'duration': 3169,
+            'duration': 198,
-            'uploader': '4v4l0n42',
+            'uploader': 'liuxt',
-            'timestamp': 1292248482.625,
+            'timestamp': 1385778501.104,
-            'upload_date': '20101213',
+            'upload_date': '20131130',
             'view_count': int,
         }
     }

youtube_dl/extractor/extractors.py

@@ -134,6 +134,7 @@ from .ccc import CCCIE
 from .cda import CDAIE
 from .ceskatelevize import CeskaTelevizeIE
 from .channel9 import Channel9IE
+from .charlierose import CharlieRoseIE
 from .chaturbate import ChaturbateIE
 from .chilloutzone import ChilloutzoneIE
 from .chirbit import (
@@ -1114,7 +1115,4 @@ from .youtube import (
 )
 from .zapiks import ZapiksIE
 from .zdf import ZDFIE, ZDFChannelIE
-from .zingmp3 import (
+from .zingmp3 import ZingMp3IE
-    ZingMp3SongIE,
-    ZingMp3AlbumIE,
-)

youtube_dl/extractor/firsttv.py

@@ -2,44 +2,40 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..compat import compat_xpath
+from ..compat import compat_urlparse
 from ..utils import (
     int_or_none,
     qualities,
     unified_strdate,
-    xpath_attr,
-    xpath_element,
-    xpath_text,
-    xpath_with_ns,
 )
 
 
 class FirstTVIE(InfoExtractor):
     IE_NAME = '1tv'
     IE_DESC = 'Первый канал'
-    _VALID_URL = r'https?://(?:www\.)?1tv\.ru/(?:[^/]+/)+p?(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:www\.)?1tv\.ru/(?:[^/]+/)+(?P<id>[^/?#]+)'
 
     _TESTS = [{
-        # single format via video_materials.json API
+        # single format
-        'url': 'http://www.1tv.ru/prj/inprivate/vypusk/35930',
+        'url': 'http://www.1tv.ru/shows/naedine-so-vsemi/vypuski/gost-lyudmila-senchina-naedine-so-vsemi-vypusk-ot-12-02-2015',
-        'md5': '82a2777648acae812d58b3f5bd42882b',
+        'md5': 'a1b6b60d530ebcf8daacf4565762bbaf',
         'info_dict': {
-            'id': '35930',
+            'id': '40049',
             'ext': 'mp4',
             'title': 'Гость Людмила Сенчина. Наедине со всеми. Выпуск от 12.02.2015',
-            'description': 'md5:357933adeede13b202c7c21f91b871b2',
+            'description': 'md5:36a39c1d19618fec57d12efe212a8370',
             'thumbnail': 're:^https?://.*\.(?:jpg|JPG)$',
             'upload_date': '20150212',
             'duration': 2694,
         },
     }, {
-        # multiple formats via video_materials.json API
+        # multiple formats
-        'url': 'http://www.1tv.ru/video_archive/projects/dobroeutro/p113641',
+        'url': 'http://www.1tv.ru/shows/dobroe-utro/pro-zdorove/vesennyaya-allergiya-dobroe-utro-fragment-vypuska-ot-07042016',
         'info_dict': {
-            'id': '113641',
+            'id': '364746',
             'ext': 'mp4',
             'title': 'Весенняя аллергия. Доброе утро. Фрагмент выпуска от 07.04.2016',
-            'description': 'md5:8dcebb3dded0ff20fade39087fd1fee2',
+            'description': 'md5:a242eea0031fd180a4497d52640a9572',
             'thumbnail': 're:^https?://.*\.(?:jpg|JPG)$',
             'upload_date': '20160407',
             'duration': 179,
@@ -48,84 +44,47 @@ class FirstTVIE(InfoExtractor):
         'params': {
             'skip_download': True,
         },
-    }, {
-        # single format only available via ONE_ONLINE_VIDEOS.archive_single_xml API
-        'url': 'http://www.1tv.ru/video_archive/series/f7552/p47038',
-        'md5': '519d306c5b5669761fd8906c39dbee23',
-        'info_dict': {
-            'id': '47038',
-            'ext': 'mp4',
-            'title': '"Побег". Второй сезон. 3 серия',
-            'description': 'md5:3abf8f6b9bce88201c33e9a3d794a00b',
-            'thumbnail': 're:^https?://.*\.(?:jpg|JPG)$',
-            'upload_date': '20120516',
-            'duration': 3080,
-        },
-    }, {
-        'url': 'http://www.1tv.ru/videoarchive/9967',
-        'only_matching': True,
     }]
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        display_id = self._match_id(url)
 
-        # Videos with multiple formats only available via this API
+        webpage = self._download_webpage(url, display_id)
-        video = self._download_json(
+        playlist_url = compat_urlparse.urljoin(url, self._search_regex(
-            'http://www.1tv.ru/video_materials.json?legacy_id=%s' % video_id,
+            r'data-playlist-url="([^"]+)', webpage, 'playlist url'))
-            video_id, fatal=False)
-
-        description, thumbnail, upload_date, duration = [None] * 4
-
-        if video:
-            item = video[0]
-            title = item['title']
-            quality = qualities(('ld', 'sd', 'hd', ))
-            formats = [{
-                'url': f['src'],
-                'format_id': f.get('name'),
-                'quality': quality(f.get('name')),
-            } for f in item['mbr'] if f.get('src')]
-            thumbnail = item.get('poster')
-        else:
-            # Some videos are not available via video_materials.json
-            video = self._download_xml(
-                'http://www.1tv.ru/owa/win/ONE_ONLINE_VIDEOS.archive_single_xml?pid=%s' % video_id,
-                video_id)
-
-            NS_MAP = {
-                'media': 'http://search.yahoo.com/mrss/',
-            }
-
-            item = xpath_element(video, './channel/item', fatal=True)
-            title = xpath_text(item, './title', fatal=True)
-            formats = [{
-                'url': content.attrib['url'],
-            } for content in item.findall(
-                compat_xpath(xpath_with_ns('./media:content', NS_MAP))) if content.attrib.get('url')]
-            thumbnail = xpath_attr(
-                item, xpath_with_ns('./media:thumbnail', NS_MAP), 'url')
 
+        item = self._download_json(playlist_url, display_id)[0]
+        video_id = item['id']
+        quality = qualities(('ld', 'sd', 'hd', ))
+        formats = []
+        for f in item.get('mbr', []):
+            src = f.get('src')
+            if not src:
+                continue
+            fname = f.get('name')
+            formats.append({
+                'url': src,
+                'format_id': fname,
+                'quality': quality(fname),
+            })
         self._sort_formats(formats)
 
-        webpage = self._download_webpage(url, video_id, 'Downloading page', fatal=False)
+        title = self._html_search_regex(
-        if webpage:
+            (r'<div class="tv_translation">\s*<h1><a href="[^"]+">([^<]*)</a>',
-            title = self._html_search_regex(
+             r"'title'\s*:\s*'([^']+)'"),
-                (r'<div class="tv_translation">\s*<h1><a href="[^"]+">([^<]*)</a>',
+            webpage, 'title', default=None) or item['title']
-                 r"'title'\s*:\s*'([^']+)'"),
+        description = self._html_search_regex(
-                webpage, 'title', default=None) or title
+            r'<div class="descr">\s*<div>&nbsp;</div>\s*<p>([^<]*)</p></div>',
-            description = self._html_search_regex(
+            webpage, 'description', default=None) or self._html_search_meta(
-                r'<div class="descr">\s*<div>&nbsp;</div>\s*<p>([^<]*)</p></div>',
+            'description', webpage, 'description')
-                webpage, 'description', default=None) or self._html_search_meta(
+        duration = int_or_none(self._html_search_meta(
-                'description', webpage, 'description')
+            'video:duration', webpage, 'video duration', fatal=False))
-            thumbnail = thumbnail or self._og_search_thumbnail(webpage)
+        upload_date = unified_strdate(self._html_search_meta(
-            duration = int_or_none(self._html_search_meta(
+            'ya:ovs:upload_date', webpage, 'upload date', fatal=False))
-                'video:duration', webpage, 'video duration', fatal=False))
-            upload_date = unified_strdate(self._html_search_meta(
-                'ya:ovs:upload_date', webpage, 'upload date', fatal=False))
 
         return {
             'id': video_id,
-            'thumbnail': thumbnail,
+            'thumbnail': item.get('poster') or self._og_search_thumbnail(webpage),
             'title': title,
             'description': description,
             'upload_date': upload_date,

youtube_dl/extractor/globo.py

@@ -396,12 +396,12 @@ class GloboIE(InfoExtractor):
 
 
 class GloboArticleIE(InfoExtractor):
-    _VALID_URL = 'https?://.+?\.globo\.com/(?:[^/]+/)*(?P<id>[^/]+)\.html'
+    _VALID_URL = 'https?://.+?\.globo\.com/(?:[^/]+/)*(?P<id>[^/]+)(?:\.html)?'
 
     _VIDEOID_REGEXES = [
         r'\bdata-video-id=["\'](\d{7,})',
         r'\bdata-player-videosids=["\'](\d{7,})',
-        r'\bvideosIDs\s*:\s*["\'](\d{7,})',
+        r'\bvideosIDs\s*:\s*["\']?(\d{7,})',
         r'\bdata-id=["\'](\d{7,})',
         r'<div[^>]+\bid=["\'](\d{7,})',
     ]
@@ -423,6 +423,9 @@ class GloboArticleIE(InfoExtractor):
     }, {
         'url': 'http://gshow.globo.com/programas/tv-xuxa/O-Programa/noticia/2014/01/xuxa-e-junno-namoram-muuuito-em-luau-de-zeze-di-camargo-e-luciano.html',
         'only_matching': True,
+    }, {
+        'url': 'http://oglobo.globo.com/rio/a-amizade-entre-um-entregador-de-farmacia-um-piano-19946271',
+        'only_matching': True,
     }]
 
     @classmethod

youtube_dl/extractor/imdb.py

@@ -6,6 +6,7 @@ from .common import InfoExtractor
 from ..utils import (
     mimetype2ext,
     qualities,
+    remove_end,
 )
 
 
@@ -19,7 +20,7 @@ class ImdbIE(InfoExtractor):
         'info_dict': {
             'id': '2524815897',
             'ext': 'mp4',
-            'title': 'Ice Age: Continental Drift Trailer (No. 2) - IMDb',
+            'title': 'Ice Age: Continental Drift Trailer (No. 2)',
             'description': 'md5:9061c2219254e5d14e03c25c98e96a81',
         }
     }, {
@@ -83,10 +84,10 @@ class ImdbIE(InfoExtractor):
 
         return {
             'id': video_id,
-            'title': self._og_search_title(webpage),
+            'title': remove_end(self._og_search_title(webpage), ' - IMDb'),
             'formats': formats,
             'description': descr,
-            'thumbnail': format_info['slate'],
+            'thumbnail': format_info.get('slate'),
         }
 
 

youtube_dl/extractor/ivi.py

@@ -1,4 +1,4 @@
-# encoding: utf-8
+# coding: utf-8
 from __future__ import unicode_literals
 
 import re
@@ -8,7 +8,7 @@ from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
     int_or_none,
-    sanitized_Request,
+    qualities,
 )
 
 
@@ -49,11 +49,27 @@ class IviIE(InfoExtractor):
                 'thumbnail': 're:^https?://.*\.jpg$',
             },
             'skip': 'Only works from Russia',
+        },
+        {
+            # with MP4-HD720 format
+            'url': 'http://www.ivi.ru/watch/146500',
+            'md5': 'd63d35cdbfa1ea61a5eafec7cc523e1e',
+            'info_dict': {
+                'id': '146500',
+                'ext': 'mp4',
+                'title': 'Кукла',
+                'description': 'md5:ffca9372399976a2d260a407cc74cce6',
+                'duration': 5599,
+                'thumbnail': 're:^https?://.*\.jpg$',
+            },
+            'skip': 'Only works from Russia',
         }
     ]
 
     # Sorted by quality
-    _KNOWN_FORMATS = ['MP4-low-mobile', 'MP4-mobile', 'FLV-lo', 'MP4-lo', 'FLV-hi', 'MP4-hi', 'MP4-SHQ']
+    _KNOWN_FORMATS = (
+        'MP4-low-mobile', 'MP4-mobile', 'FLV-lo', 'MP4-lo', 'FLV-hi', 'MP4-hi',
+        'MP4-SHQ', 'MP4-HD720', 'MP4-HD1080')
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -69,10 +85,9 @@ class IviIE(InfoExtractor):
             ]
         }
 
-        request = sanitized_Request(
-            'http://api.digitalaccess.ru/api/json/', json.dumps(data))
         video_json = self._download_json(
-            request, video_id, 'Downloading video JSON')
+            'http://api.digitalaccess.ru/api/json/', video_id,
+            'Downloading video JSON', data=json.dumps(data))
 
         if 'error' in video_json:
             error = video_json['error']
@@ -84,11 +99,13 @@ class IviIE(InfoExtractor):
 
         result = video_json['result']
 
+        quality = qualities(self._KNOWN_FORMATS)
+
         formats = [{
             'url': x['url'],
-            'format_id': x['content_format'],
+            'format_id': x.get('content_format'),
-            'preference': self._KNOWN_FORMATS.index(x['content_format']),
+            'quality': quality(x.get('content_format')),
-        } for x in result['files'] if x['content_format'] in self._KNOWN_FORMATS]
+        } for x in result['files'] if x.get('url')]
 
         self._sort_formats(formats)
 
@@ -115,7 +132,7 @@ class IviIE(InfoExtractor):
             webpage, 'season number', default=None))
 
         episode_number = int_or_none(self._search_regex(
-            r'<meta[^>]+itemprop="episode"[^>]*>\s*<meta[^>]+itemprop="episodeNumber"[^>]+content="(\d+)',
+            r'[^>]+itemprop="episode"[^>]*>\s*<meta[^>]+itemprop="episodeNumber"[^>]+content="(\d+)',
             webpage, 'episode number', default=None))
 
         description = self._og_search_description(webpage, default=None) or self._html_search_meta(

youtube_dl/extractor/kaltura.py

@@ -36,6 +36,12 @@ class KalturaIE(InfoExtractor):
                 '''
     _SERVICE_URL = 'http://cdnapi.kaltura.com'
     _SERVICE_BASE = '/api_v3/index.php'
+    # See https://github.com/kaltura/server/blob/master/plugins/content/caption/base/lib/model/enums/CaptionType.php
+    _CAPTION_TYPES = {
+        1: 'srt',
+        2: 'ttml',
+        3: 'vtt',
+    }
     _TESTS = [
         {
             'url': 'kaltura:269692:1_1jc2y3e4',
@@ -67,6 +73,27 @@ class KalturaIE(InfoExtractor):
             # video with subtitles
             'url': 'kaltura:111032:1_cw786r8q',
             'only_matching': True,
+        },
+        {
+            # video with ttml subtitles (no fileExt)
+            'url': 'kaltura:1926081:0_l5ye1133',
+            'info_dict': {
+                'id': '0_l5ye1133',
+                'ext': 'mp4',
+                'title': 'What Can You Do With Python?',
+                'upload_date': '20160221',
+                'uploader_id': 'stork',
+                'thumbnail': 're:^https?://.*/thumbnail/.*',
+                'timestamp': int,
+                'subtitles': {
+                    'en': [{
+                        'ext': 'ttml',
+                    }],
+                },
+            },
+            'params': {
+                'skip_download': True,
+            },
         }
     ]
 
@@ -122,18 +149,6 @@ class KalturaIE(InfoExtractor):
 
         return data
 
-    def _get_kaltura_signature(self, video_id, partner_id, service_url=None):
-        actions = [{
-            'apiVersion': '3.1',
-            'expiry': 86400,
-            'format': 1,
-            'service': 'session',
-            'action': 'startWidgetSession',
-            'widgetId': '_%s' % partner_id,
-        }]
-        return self._kaltura_api_call(
-            video_id, actions, service_url, note='Downloading Kaltura signature')['ks']
-
     def _get_video_info(self, video_id, partner_id, service_url=None):
         actions = [
             {
@@ -208,6 +223,17 @@ class KalturaIE(InfoExtractor):
                     reference_id)['entryResult']
                 info, flavor_assets = entry_data['meta'], entry_data['contextData']['flavorAssets']
                 entry_id = info['id']
+                # Unfortunately, data returned in kalturaIframePackageData lacks
+                # captions so we will try requesting the complete data using
+                # regular approach since we now know the entry_id
+                try:
+                    _, info, flavor_assets, captions = self._get_video_info(
+                        entry_id, partner_id)
+                except ExtractorError:
+                    # Regular scenario failed but we already have everything
+                    # extracted apart from captions and can process at least
+                    # with this
+                    pass
             else:
                 raise ExtractorError('Invalid URL', expected=True)
             ks = params.get('flashvars[ks]', [None])[0]
@@ -265,9 +291,12 @@ class KalturaIE(InfoExtractor):
                 # Continue if caption is not ready
                 if f.get('status') != 2:
                     continue
+                if not caption.get('id'):
+                    continue
+                caption_format = int_or_none(caption.get('format'))
                 subtitles.setdefault(caption.get('languageCode') or caption.get('language'), []).append({
                     'url': '%s/api_v3/service/caption_captionasset/action/serve/captionAssetId/%s' % (self._SERVICE_URL, caption['id']),
-                    'ext': caption.get('fileExt'),
+                    'ext': caption.get('fileExt') or self._CAPTION_TYPES.get(caption_format) or 'ttml',
                 })
 
         return {

youtube_dl/extractor/litv.py

@@ -14,7 +14,7 @@ from ..utils import (
 
 
 class LiTVIE(InfoExtractor):
-    _VALID_URL = r'https?://www\.litv\.tv/vod/[^/]+/content\.do\?.*?\bid=(?P<id>[^&]+)'
+    _VALID_URL = r'https?://www\.litv\.tv/(?:vod|promo)/[^/]+/(?:content\.do)?\?.*?\b(?:content_)?id=(?P<id>[^&]+)'
 
     _URL_TEMPLATE = 'https://www.litv.tv/vod/%s/content.do?id=%s'
 
@@ -27,6 +27,7 @@ class LiTVIE(InfoExtractor):
         'playlist_count': 50,
     }, {
         'url': 'https://www.litv.tv/vod/drama/content.do?brc_id=root&id=VOD00041610&isUHEnabled=true&autoPlay=1',
+        'md5': '969e343d9244778cb29acec608e53640',
         'info_dict': {
             'id': 'VOD00041610',
             'ext': 'mp4',
@@ -37,7 +38,16 @@ class LiTVIE(InfoExtractor):
         },
         'params': {
             'noplaylist': True,
-            'skip_download': True,  # m3u8 download
+        },
+        'skip': 'Georestricted to Taiwan',
+    }, {
+        'url': 'https://www.litv.tv/promo/miyuezhuan/?content_id=VOD00044841&',
+        'md5': '88322ea132f848d6e3e18b32a832b918',
+        'info_dict': {
+            'id': 'VOD00044841',
+            'ext': 'mp4',
+            'title': '芈月傳第1集　霸星芈月降世楚國',
+            'description': '楚威王二年，太史令唐昧夜觀星象，發現霸星即將現世。王后得知霸星的預言後，想盡辦法不讓孩子順利出生，幸得莒姬相護化解危機。沒想到眾人期待下出生的霸星卻是位公主，楚威王對此失望至極。楚王后命人將女嬰丟棄河中，居然奇蹟似的被少司命像攔下，楚威王認為此女非同凡響，為她取名芈月。',
         },
         'skip': 'Georestricted to Taiwan',
     }]
@@ -92,13 +102,18 @@ class LiTVIE(InfoExtractor):
         # endpoint gives the same result as the data embedded in the webpage.
         # If georestricted, there are no embedded data, so an extra request is
         # necessary to get the error code
+        if 'assetId' not in view_data:
+            view_data = self._download_json(
+                'https://www.litv.tv/vod/ajax/getProgramInfo', video_id,
+                query={'contentId': video_id},
+                headers={'Accept': 'application/json'})
         video_data = self._parse_json(self._search_regex(
             r'uiHlsUrl\s*=\s*testBackendData\(([^;]+)\);',
             webpage, 'video data', default='{}'), video_id)
         if not video_data:
             payload = {
                 'assetId': view_data['assetId'],
-                'watchDevices': vod_data['watchDevices'],
+                'watchDevices': view_data['watchDevices'],
                 'contentType': view_data['contentType'],
             }
             video_data = self._download_json(
@@ -115,7 +130,8 @@ class LiTVIE(InfoExtractor):
             raise ExtractorError('Unexpected result from %s' % self.IE_NAME)
 
         formats = self._extract_m3u8_formats(
-            video_data['fullpath'], video_id, ext='mp4', m3u8_id='hls')
+            video_data['fullpath'], video_id, ext='mp4',
+            entry_protocol='m3u8_native', m3u8_id='hls')
         for a_format in formats:
             # LiTV HLS segments doesn't like compressions
             a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = True

youtube_dl/extractor/miomio.py

@@ -25,10 +25,7 @@ class MioMioIE(InfoExtractor):
             'title': '【SKY】字幕 铠武昭和VS平成 假面骑士大战FEAT战队 魔星字幕组 字幕',
             'duration': 5923,
         },
-        'params': {
+        'skip': 'Unable to load videos',
-            # The server provides broken file
-            'skip_download': True,
-        }
     }, {
         'url': 'http://www.miomio.tv/watch/cc184024/',
         'info_dict': {
@@ -47,16 +44,12 @@ class MioMioIE(InfoExtractor):
         'skip': 'Unable to load videos',
     }, {
         # new 'h5' player
-        'url': 'http://www.miomio.tv/watch/cc273295/',
+        'url': 'http://www.miomio.tv/watch/cc273997/',
-        'md5': '',
+        'md5': '0b27a4b4495055d826813f8c3a6b2070',
         'info_dict': {
-            'id': '273295',
+            'id': '273997',
             'ext': 'mp4',
-            'title': 'アウト×デラックス 20160526',
+            'title': 'マツコの知らない世界【劇的進化SP！ビニール傘＆冷凍食品2016】 1_2 - 16 05 31',
-        },
-        'params': {
-            # intermittent HTTP 500
-            'skip_download': True,
         },
     }]
 
@@ -116,7 +109,7 @@ class MioMioIE(InfoExtractor):
             player_webpage = self._download_webpage(
                 player_url, video_id,
                 note='Downloading player webpage', headers={'Referer': url})
-            entries = self._parse_html5_media_entries(player_url, player_webpage)
+            entries = self._parse_html5_media_entries(player_url, player_webpage, video_id)
             http_headers = {'Referer': player_url}
         else:
             http_headers = {'Referer': 'http://www.miomio.tv%s' % mioplayer_path}

youtube_dl/extractor/ntvde.py

@@ -1,6 +1,8 @@
-# encoding: utf-8
+# coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 from ..compat import compat_urlparse
 from ..utils import (
@@ -40,8 +42,8 @@ class NTVDeIE(InfoExtractor):
         timestamp = int_or_none(info.get('publishedDateAsUnixTimeStamp'))
         vdata = self._parse_json(self._search_regex(
             r'(?s)\$\(\s*"\#player"\s*\)\s*\.data\(\s*"player",\s*(\{.*?\})\);',
-            webpage, 'player data'),
+            webpage, 'player data'), video_id,
-            video_id, transform_source=js_to_json)
+            transform_source=lambda s: js_to_json(re.sub(r'advertising:\s*{[^}]+},', '', s)))
         duration = parse_duration(vdata.get('duration'))
 
         formats = []

youtube_dl/extractor/radiobremen.py

@@ -13,15 +13,15 @@ class RadioBremenIE(InfoExtractor):
     IE_NAME = 'radiobremen'
 
     _TEST = {
-        'url': 'http://www.radiobremen.de/mediathek/index.html?id=114720',
+        'url': 'http://www.radiobremen.de/mediathek/?id=141876',
         'info_dict': {
-            'id': '114720',
+            'id': '141876',
             'ext': 'mp4',
-            'duration': 1685,
+            'duration': 178,
             'width': 512,
-            'title': 'buten un binnen vom 22. Dezember',
+            'title': 'Druck auf Patrick Öztürk',
             'thumbnail': 're:https?://.*\.jpg$',
-            'description': 'Unter anderem mit diesen Themen: 45 Flüchtlinge sind in Worpswede angekommen +++ Freies Internet für alle: Bremer arbeiten an einem flächendeckenden W-Lan-Netzwerk +++ Aktivisten kämpfen für das Unibad +++ So war das Wetter 2014 +++',
+            'description': 'Gegen den SPD-Bürgerschaftsabgeordneten Patrick Öztürk wird wegen Beihilfe zum gewerbsmäßigen Betrug ermittelt. Am Donnerstagabend sollte er dem Vorstand des SPD-Unterbezirks Bremerhaven dazu Rede und Antwort stehen.',
         },
     }
 

youtube_dl/extractor/snotr.py

@@ -5,9 +5,9 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
-    float_or_none,
-    str_to_int,
     parse_duration,
+    parse_filesize,
+    str_to_int,
 )
 
 
@@ -17,21 +17,24 @@ class SnotrIE(InfoExtractor):
         'url': 'http://www.snotr.com/video/13708/Drone_flying_through_fireworks',
         'info_dict': {
             'id': '13708',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'Drone flying through fireworks!',
-            'duration': 247,
+            'duration': 248,
-            'filesize_approx': 98566144,
+            'filesize_approx': 40700000,
             'description': 'A drone flying through Fourth of July Fireworks',
-        }
+            'thumbnail': 're:^https?://.*\.jpg$',
+        },
+        'expected_warnings': ['description'],
     }, {
         'url': 'http://www.snotr.com/video/530/David_Letteman_-_George_W_Bush_Top_10',
         'info_dict': {
             'id': '530',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'David Letteman - George W. Bush Top 10',
             'duration': 126,
-            'filesize_approx': 8912896,
+            'filesize_approx': 8500000,
             'description': 'The top 10 George W. Bush moments, brought to you by David Letterman!',
+            'thumbnail': 're:^https?://.*\.jpg$',
         }
     }]
 
@@ -43,26 +46,28 @@ class SnotrIE(InfoExtractor):
         title = self._og_search_title(webpage)
 
         description = self._og_search_description(webpage)
-        video_url = 'http://cdn.videos.snotr.com/%s.flv' % video_id
+        info_dict = self._parse_html5_media_entries(
+            url, webpage, video_id, m3u8_entry_protocol='m3u8_native')[0]
 
         view_count = str_to_int(self._html_search_regex(
-            r'<p>\n<strong>Views:</strong>\n([\d,\.]+)</p>',
+            r'<p[^>]*>\s*<strong[^>]*>Views:</strong>\s*<span[^>]*>([\d,\.]+)',
             webpage, 'view count', fatal=False))
 
         duration = parse_duration(self._html_search_regex(
-            r'<p>\n<strong>Length:</strong>\n\s*([0-9:]+).*?</p>',
+            r'<p[^>]*>\s*<strong[^>]*>Length:</strong>\s*<span[^>]*>([\d:]+)',
             webpage, 'duration', fatal=False))
 
-        filesize_approx = float_or_none(self._html_search_regex(
+        filesize_approx = parse_filesize(self._html_search_regex(
-            r'<p>\n<strong>Filesize:</strong>\n\s*([0-9.]+)\s*megabyte</p>',
+            r'<p[^>]*>\s*<strong[^>]*>Filesize:</strong>\s*<span[^>]*>([^<]+)',
-            webpage, 'filesize', fatal=False), invscale=1024 * 1024)
+            webpage, 'filesize', fatal=False))
 
-        return {
+        info_dict.update({
             'id': video_id,
             'description': description,
             'title': title,
-            'url': video_url,
             'view_count': view_count,
             'duration': duration,
             'filesize_approx': filesize_approx,
-        }
+        })
+
+        return info_dict

youtube_dl/extractor/twitch.py

@@ -7,6 +7,7 @@ import random
 
 from .common import InfoExtractor
 from ..compat import (
+    compat_HTTPError,
     compat_parse_qs,
     compat_str,
     compat_urllib_parse_urlencode,
@@ -14,13 +15,13 @@ from ..compat import (
     compat_urlparse,
 )
 from ..utils import (
+    clean_html,
     ExtractorError,
     int_or_none,
     js_to_json,
     orderedSet,
     parse_duration,
     parse_iso8601,
-    sanitized_Request,
     urlencode_postdata,
 )
 
@@ -42,7 +43,7 @@ class TwitchBaseIE(InfoExtractor):
                 '%s returned error: %s - %s' % (self.IE_NAME, error, response.get('message')),
                 expected=True)
 
-    def _download_json(self, url, video_id, note='Downloading JSON metadata'):
+    def _call_api(self, path, item_id, note):
         headers = {
             'Referer': 'http://api.twitch.tv/crossdomain/receiver.html?v=2',
             'X-Requested-With': 'XMLHttpRequest',
@@ -50,8 +51,8 @@ class TwitchBaseIE(InfoExtractor):
         for cookie in self._downloader.cookiejar:
             if cookie.name == 'api_token':
                 headers['Twitch-Api-Token'] = cookie.value
-        request = sanitized_Request(url, headers=headers)
+        response = self._download_json(
-        response = super(TwitchBaseIE, self)._download_json(request, video_id, note)
+            '%s/%s' % (self._API_BASE, path), item_id, note)
         self._handle_error(response)
         return response
 
@@ -63,9 +64,17 @@ class TwitchBaseIE(InfoExtractor):
         if username is None:
             return
 
+        def fail(message):
+            raise ExtractorError(
+                'Unable to login. Twitch said: %s' % message, expected=True)
+
         login_page, handle = self._download_webpage_handle(
             self._LOGIN_URL, None, 'Downloading login page')
 
+        # Some TOR nodes and public proxies are blocked completely
+        if 'blacklist_message' in login_page:
+            fail(clean_html(login_page))
+
         login_form = self._hidden_inputs(login_page)
 
         login_form.update({
@@ -82,21 +91,24 @@ class TwitchBaseIE(InfoExtractor):
         if not post_url.startswith('http'):
             post_url = compat_urlparse.urljoin(redirect_url, post_url)
 
-        request = sanitized_Request(
+        headers = {'Referer': redirect_url}
-            post_url, urlencode_postdata(login_form))
-        request.add_header('Referer', redirect_url)
-        response = self._download_webpage(
-            request, None, 'Logging in as %s' % username)
 
-        error_message = self._search_regex(
+        try:
-            r'<div[^>]+class="subwindow_notice"[^>]*>([^<]+)</div>',
+            response = self._download_json(
-            response, 'error message', default=None)
+                post_url, None, 'Logging in as %s' % username,
-        if error_message:
+                data=urlencode_postdata(login_form),
-            raise ExtractorError(
+                headers=headers)
-                'Unable to login. Twitch said: %s' % error_message, expected=True)
+        except ExtractorError as e:
+            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 400:
+                response = self._parse_json(
+                    e.cause.read().decode('utf-8'), None)
+                fail(response['message'])
+            raise
 
-        if '>Reset your password<' in response:
+        if response.get('redirect'):
-            self.report_warning('Twitch asks you to reset your password, go to https://secure.twitch.tv/reset/submit')
+            self._download_webpage(
+                response['redirect'], None, 'Downloading login redirect page',
+                headers=headers)
 
     def _prefer_source(self, formats):
         try:
@@ -109,14 +121,14 @@ class TwitchBaseIE(InfoExtractor):
 
 class TwitchItemBaseIE(TwitchBaseIE):
     def _download_info(self, item, item_id):
-        return self._extract_info(self._download_json(
+        return self._extract_info(self._call_api(
-            '%s/kraken/videos/%s%s' % (self._API_BASE, item, item_id), item_id,
+            'kraken/videos/%s%s' % (item, item_id), item_id,
             'Downloading %s info JSON' % self._ITEM_TYPE))
 
     def _extract_media(self, item_id):
         info = self._download_info(self._ITEM_SHORTCUT, item_id)
-        response = self._download_json(
+        response = self._call_api(
-            '%s/api/videos/%s%s' % (self._API_BASE, self._ITEM_SHORTCUT, item_id), item_id,
+            'api/videos/%s%s' % (self._ITEM_SHORTCUT, item_id), item_id,
             'Downloading %s playlist JSON' % self._ITEM_TYPE)
         entries = []
         chunks = response['chunks']
@@ -246,8 +258,8 @@ class TwitchVodIE(TwitchItemBaseIE):
         item_id = self._match_id(url)
 
         info = self._download_info(self._ITEM_SHORTCUT, item_id)
-        access_token = self._download_json(
+        access_token = self._call_api(
-            '%s/api/vods/%s/access_token' % (self._API_BASE, item_id), item_id,
+            'api/vods/%s/access_token' % item_id, item_id,
             'Downloading %s access token' % self._ITEM_TYPE)
 
         formats = self._extract_m3u8_formats(
@@ -275,12 +287,12 @@ class TwitchVodIE(TwitchItemBaseIE):
 
 
 class TwitchPlaylistBaseIE(TwitchBaseIE):
-    _PLAYLIST_URL = '%s/kraken/channels/%%s/videos/?offset=%%d&limit=%%d' % TwitchBaseIE._API_BASE
+    _PLAYLIST_PATH = 'kraken/channels/%s/videos/?offset=%d&limit=%d'
     _PAGE_LIMIT = 100
 
     def _extract_playlist(self, channel_id):
-        info = self._download_json(
+        info = self._call_api(
-            '%s/kraken/channels/%s' % (self._API_BASE, channel_id),
+            'kraken/channels/%s' % channel_id,
             channel_id, 'Downloading channel info JSON')
         channel_name = info.get('display_name') or info.get('name')
         entries = []
@@ -289,8 +301,8 @@ class TwitchPlaylistBaseIE(TwitchBaseIE):
         broken_paging_detected = False
         counter_override = None
         for counter in itertools.count(1):
-            response = self._download_json(
+            response = self._call_api(
-                self._PLAYLIST_URL % (channel_id, offset, limit),
+                self._PLAYLIST_PATH % (channel_id, offset, limit),
                 channel_id,
                 'Downloading %s videos JSON page %s'
                 % (self._PLAYLIST_TYPE, counter_override or counter))
@@ -345,7 +357,7 @@ class TwitchProfileIE(TwitchPlaylistBaseIE):
 class TwitchPastBroadcastsIE(TwitchPlaylistBaseIE):
     IE_NAME = 'twitch:past_broadcasts'
     _VALID_URL = r'%s/(?P<id>[^/]+)/profile/past_broadcasts/?(?:\#.*)?$' % TwitchBaseIE._VALID_URL_BASE
-    _PLAYLIST_URL = TwitchPlaylistBaseIE._PLAYLIST_URL + '&broadcasts=true'
+    _PLAYLIST_PATH = TwitchPlaylistBaseIE._PLAYLIST_PATH + '&broadcasts=true'
     _PLAYLIST_TYPE = 'past broadcasts'
 
     _TEST = {
@@ -389,8 +401,8 @@ class TwitchStreamIE(TwitchBaseIE):
     def _real_extract(self, url):
         channel_id = self._match_id(url)
 
-        stream = self._download_json(
+        stream = self._call_api(
-            '%s/kraken/streams/%s' % (self._API_BASE, channel_id), channel_id,
+            'kraken/streams/%s' % channel_id, channel_id,
             'Downloading stream JSON').get('stream')
 
         # Fallback on profile extraction if stream is offline
@@ -405,8 +417,8 @@ class TwitchStreamIE(TwitchBaseIE):
         # JSON and fallback to lowercase if it's not available.
         channel_id = stream.get('channel', {}).get('name') or channel_id.lower()
 
-        access_token = self._download_json(
+        access_token = self._call_api(
-            '%s/api/channels/%s/access_token' % (self._API_BASE, channel_id), channel_id,
+            'api/channels/%s/access_token' % channel_id, channel_id,
             'Downloading channel access token')
 
         query = {

youtube_dl/extractor/vuclip.py

@@ -17,12 +17,12 @@ class VuClipIE(InfoExtractor):
     _VALID_URL = r'https?://(?:m\.)?vuclip\.com/w\?.*?cid=(?P<id>[0-9]+)'
 
     _TEST = {
-        'url': 'http://m.vuclip.com/w?cid=922692425&fid=70295&z=1010&nvar&frm=index.html',
+        'url': 'http://m.vuclip.com/w?cid=1129900602&bu=8589892792&frm=w&z=34801&op=0&oc=843169247&section=recommend',
         'info_dict': {
-            'id': '922692425',
+            'id': '1129900602',
             'ext': '3gp',
-            'title': 'The Toy Soldiers - Hollywood Movie Trailer',
+            'title': 'Top 10 TV Convicts',
-            'duration': 177,
+            'duration': 733,
         }
     }
 
@@ -54,7 +54,7 @@ class VuClipIE(InfoExtractor):
                 'url': video_url,
             }]
         else:
-            formats = self._parse_html5_media_entries(url, webpage)[0]['formats']
+            formats = self._parse_html5_media_entries(url, webpage, video_id)[0]['formats']
 
         title = remove_end(self._html_search_regex(
             r'<title>(.*?)-\s*Vuclip</title>', webpage, 'title').strip(), ' - Video')

youtube_dl/extractor/zingmp3.py

@@ -4,13 +4,17 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import ExtractorError
+from ..utils import (
+    ExtractorError,
+    int_or_none,
+    update_url_query,
+)
 
 
 class ZingMp3BaseInfoExtractor(InfoExtractor):
 
-    def _extract_item(self, item, fatal=True):
+    def _extract_item(self, item, page_type, fatal=True):
-        error_message = item.find('./errormessage').text
+        error_message = item.get('msg')
         if error_message:
             if not fatal:
                 return
@@ -18,25 +22,48 @@ class ZingMp3BaseInfoExtractor(InfoExtractor):
                 '%s returned error: %s' % (self.IE_NAME, error_message),
                 expected=True)
 
-        title = item.find('./title').text.strip()
+        formats = []
-        source = item.find('./source').text
+        for quality, source_url in zip(item.get('qualities') or item.get('quality', []), item.get('source_list') or item.get('source', [])):
-        extension = item.attrib['type']
+            if not source_url or source_url == 'require vip':
-        thumbnail = item.find('./backimage').text
+                continue
+            if not re.match(r'https?://', source_url):
+                source_url = '//' + source_url
+            source_url = self._proto_relative_url(source_url, 'http:')
+            quality_num = int_or_none(quality)
+            f = {
+                'format_id': quality,
+                'url': source_url,
+            }
+            if page_type == 'video':
+                f.update({
+                    'height': quality_num,
+                    'ext': 'mp4',
+                })
+            else:
+                f.update({
+                    'abr': quality_num,
+                    'ext': 'mp3',
+                })
+            formats.append(f)
+
+        cover = item.get('cover')
 
         return {
-            'title': title,
+            'title': (item.get('name') or item.get('title')).strip(),
-            'url': source,
+            'formats': formats,
-            'ext': extension,
+            'thumbnail': 'http:/' + cover if cover else None,
-            'thumbnail': thumbnail,
+            'artist': item.get('artist'),
         }
 
-    def _extract_player_xml(self, player_xml_url, id, playlist_title=None):
+    def _extract_player_json(self, player_json_url, id, page_type, playlist_title=None):
-        player_xml = self._download_xml(player_xml_url, id, 'Downloading Player XML')
+        player_json = self._download_json(player_json_url, id, 'Downloading Player JSON')
-        items = player_xml.findall('./item')
+        items = player_json['data']
+        if 'item' in items:
+            items = items['item']
 
         if len(items) == 1:
             # one single song
-            data = self._extract_item(items[0])
+            data = self._extract_item(items[0], page_type)
             data['id'] = id
 
             return data
@@ -45,7 +72,7 @@ class ZingMp3BaseInfoExtractor(InfoExtractor):
             entries = []
 
             for i, item in enumerate(items, 1):
-                entry = self._extract_item(item, fatal=False)
+                entry = self._extract_item(item, page_type, fatal=False)
                 if not entry:
                     continue
                 entry['id'] = '%s-%d' % (id, i)
@@ -59,8 +86,8 @@ class ZingMp3BaseInfoExtractor(InfoExtractor):
             }
 
 
-class ZingMp3SongIE(ZingMp3BaseInfoExtractor):
+class ZingMp3IE(ZingMp3BaseInfoExtractor):
-    _VALID_URL = r'https?://mp3\.zing\.vn/bai-hat/(?P<slug>[^/]+)/(?P<song_id>\w+)\.html'
+    _VALID_URL = r'https?://mp3\.zing\.vn/(?:bai-hat|album|playlist|video-clip)/[^/]+/(?P<id>\w+)\.html'
     _TESTS = [{
         'url': 'http://mp3.zing.vn/bai-hat/Xa-Mai-Xa-Bao-Thy/ZWZB9WAB.html',
         'md5': 'ead7ae13693b3205cbc89536a077daed',
@@ -70,51 +97,47 @@ class ZingMp3SongIE(ZingMp3BaseInfoExtractor):
             'ext': 'mp3',
             'thumbnail': 're:^https?://.*\.jpg$',
         },
-    }]
+    }, {
-    IE_NAME = 'zingmp3:song'
+        'url': 'http://mp3.zing.vn/video-clip/Let-It-Go-Frozen-OST-Sungha-Jung/ZW6BAEA0.html',
-    IE_DESC = 'mp3.zing.vn songs'
+        'md5': '870295a9cd8045c0e15663565902618d',
-
+        'info_dict': {
-    def _real_extract(self, url):
+            'id': 'ZW6BAEA0',
-        matched = re.match(self._VALID_URL, url)
+            'title': 'Let It Go (Frozen OST)',
-        slug = matched.group('slug')
+            'ext': 'mp4',
-        song_id = matched.group('song_id')
+        },
-
+    }, {
-        webpage = self._download_webpage(
-            'http://mp3.zing.vn/bai-hat/%s/%s.html' % (slug, song_id), song_id)
-
-        player_xml_url = self._search_regex(
-            r'&amp;xmlURL=(?P<xml_url>[^&]+)&', webpage, 'player xml url')
-
-        return self._extract_player_xml(player_xml_url, song_id)
-
-
-class ZingMp3AlbumIE(ZingMp3BaseInfoExtractor):
-    _VALID_URL = r'https?://mp3\.zing\.vn/(?:album|playlist)/(?P<slug>[^/]+)/(?P<album_id>\w+)\.html'
-    _TESTS = [{
         'url': 'http://mp3.zing.vn/album/Lau-Dai-Tinh-Ai-Bang-Kieu-Minh-Tuyet/ZWZBWDAF.html',
         'info_dict': {
             '_type': 'playlist',
             'id': 'ZWZBWDAF',
-            'title': 'Lâu Đài Tình Ái - Bằng Kiều ft. Minh Tuyết | Album 320 lossless',
+            'title': 'Lâu Đài Tình Ái - Bằng Kiều,Minh Tuyết | Album 320 lossless',
         },
         'playlist_count': 10,
+        'skip': 'removed at the request of the owner',
     }, {
         'url': 'http://mp3.zing.vn/playlist/Duong-Hong-Loan-apollobee/IWCAACCB.html',
         'only_matching': True,
     }]
-    IE_NAME = 'zingmp3:album'
+    IE_NAME = 'zingmp3'
-    IE_DESC = 'mp3.zing.vn albums'
+    IE_DESC = 'mp3.zing.vn'
 
     def _real_extract(self, url):
-        matched = re.match(self._VALID_URL, url)
+        page_id = self._match_id(url)
-        slug = matched.group('slug')
-        album_id = matched.group('album_id')
 
-        webpage = self._download_webpage(
+        webpage = self._download_webpage(url, page_id)
-            'http://mp3.zing.vn/album/%s/%s.html' % (slug, album_id), album_id)
-        player_xml_url = self._search_regex(
-            r'&amp;xmlURL=(?P<xml_url>[^&]+)&', webpage, 'player xml url')
 
-        return self._extract_player_xml(
+        player_json_url = self._search_regex([
-            player_xml_url, album_id,
+            r'data-xml="([^"]+)',
-            playlist_title=self._og_search_title(webpage))
+            r'&amp;xmlURL=([^&]+)&'
+        ], webpage, 'player xml url')
+
+        playlist_title = None
+        page_type = self._search_regex(r'/(?:html5)?xml/([^/-]+)', player_json_url, 'page type')
+        if page_type == 'video':
+            player_json_url = update_url_query(player_json_url, {'format': 'json'})
+        else:
+            player_json_url = player_json_url.replace('/xml/', '/html5xml/')
+            if page_type == 'album':
+                playlist_title = self._og_search_title(webpage)
+
+        return self._extract_player_json(player_json_url, page_id, page_type, playlist_title)

youtube_dl/utils.py

@@ -1504,46 +1504,63 @@ def parse_filesize(s):
     _UNIT_TABLE = {
         'B': 1,
         'b': 1,
+        'bytes': 1,
         'KiB': 1024,
         'KB': 1000,
         'kB': 1024,
         'Kb': 1000,
         'kb': 1000,
+        'kilobytes': 1000,
+        'kibibytes': 1024,
         'MiB': 1024 ** 2,
         'MB': 1000 ** 2,
         'mB': 1024 ** 2,
         'Mb': 1000 ** 2,
         'mb': 1000 ** 2,
+        'megabytes': 1000 ** 2,
+        'mebibytes': 1024 ** 2,
         'GiB': 1024 ** 3,
         'GB': 1000 ** 3,
         'gB': 1024 ** 3,
         'Gb': 1000 ** 3,
         'gb': 1000 ** 3,
+        'gigabytes': 1000 ** 3,
+        'gibibytes': 1024 ** 3,
         'TiB': 1024 ** 4,
         'TB': 1000 ** 4,
         'tB': 1024 ** 4,
         'Tb': 1000 ** 4,
         'tb': 1000 ** 4,
+        'terabytes': 1000 ** 4,
+        'tebibytes': 1024 ** 4,
         'PiB': 1024 ** 5,
         'PB': 1000 ** 5,
         'pB': 1024 ** 5,
         'Pb': 1000 ** 5,
         'pb': 1000 ** 5,
+        'petabytes': 1000 ** 5,
+        'pebibytes': 1024 ** 5,
         'EiB': 1024 ** 6,
         'EB': 1000 ** 6,
         'eB': 1024 ** 6,
         'Eb': 1000 ** 6,
         'eb': 1000 ** 6,
+        'exabytes': 1000 ** 6,
+        'exbibytes': 1024 ** 6,
         'ZiB': 1024 ** 7,
         'ZB': 1000 ** 7,
         'zB': 1024 ** 7,
         'Zb': 1000 ** 7,
         'zb': 1000 ** 7,
+        'zettabytes': 1000 ** 7,
+        'zebibytes': 1024 ** 7,
         'YiB': 1024 ** 8,
         'YB': 1000 ** 8,
         'yB': 1024 ** 8,
         'Yb': 1000 ** 8,
         'yb': 1000 ** 8,
+        'yottabytes': 1000 ** 8,
+        'yobibytes': 1024 ** 8,
     }
 
     return lookup_unit_table(_UNIT_TABLE, s)
@@ -2038,14 +2055,14 @@ def js_to_json(code):
             }.get(m.group(0), m.group(0)), v[1:-1])
 
         INTEGER_TABLE = (
-            (r'^0[xX][0-9a-fA-F]+', 16),
+            (r'^(0[xX][0-9a-fA-F]+)\s*:?$', 16),
-            (r'^0+[0-7]+', 8),
+            (r'^(0+[0-7]+)\s*:?$', 8),
         )
 
         for regex, base in INTEGER_TABLE:
             im = re.match(regex, v)
             if im:
-                i = int(im.group(0), base)
+                i = int(im.group(1), base)
                 return '"%d":' % i if v.endswith(':') else '%d' % i
 
         return '"%s"' % v

youtube_dl/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2016.08.19'
+__version__ = '2016.08.22'